AAV9-Mediated Cdk5 Inhibitory Peptide Reduces Hyperphosphorylated Tau and Inflammation and Ameliorates Behavioral Changes Caused by Overexpression of p25 in the Brain

Abstract

Background:

Under stress stimulation, p25 is generated by cleavage of p35 and acts as an activator of cyclin-dependent kinase 5 (Cdk5) like p35. Unlike Cdk5/p35, which is important for brain development, aberrant activity of Cdk5/p25 plays a pathological role in neurodegenerative diseases, such as Alzheimer’s disease, by inducing hyperphosphorylation of downstream substrates related to pathological progression. A truncated fragment of the c-terminus of p35, the Cdk5 inhibitory peptide (CIP), selectively inhibits Cdk5/ p25 activity in cultured neurons and in CIP/p25 tetra-transgenic mice.

Objective:

First, we aimed to establish a p25 overexpression adult mouse model, then to evaluate whether CIP delivered by adeno-associated virus serotype 9 (AAV9) can ameliorate neuronal toxicity induced by p25.

Methods:

The p25 overexpression mouse model was established by intracerebroventricular (i.c.v.) injection of AAV8-GFP-p25 in 8-week-old mice. One month later, these mice were i.c.v. injected with AAV9-CIP-T2A-mCherry or AAV9 vector as control. Pathological and behavioral changes were assessed 3-months post-injection in all mice.

Results:

The p25 overexpression mice displayed hyperphosphorylation of tau at multiple sites, activation of astrocytes, and elevated inflammatory factors, including IL-1 and TNF-α, which were significantly decreased by the administration of CIP. However, Aβ deposition and microgliosis were not obvious in p25 overexpression mice. In addition, a significant learning decline and anxiety-like behavior were induced by p25 toxicity, and CIP treatment improved learning ability in p25 mice.

Conclusion:

AAV-mediated p25 overexpression mouse model is easy to construct to study p25-induced neuronal toxicity. Application of CIP after p25 insult reverses the pathological changes and behavioral abnormalities.

Keywords

Adeno-associated virus Cdk5 inhibitory peptide hyperphosphorylation of tau neurodegeneration p25

INTRODUCTION

Cyclin-dependent kinase 5 (Cdk5) is a member of the Cdk family of serine/threonine kinases [1]. Monomeric Cdk5 displays no enzymatic activity [2], but neuronal specific regulatory protein p35 or p39 activates Cdk5, which plays a critical role in sustaining a broad range of neuronal functions, such as neuronal migration, synaptogenesis, cellular adhesion, and neurite outgrowth [3 –5].

The calpain cleavage product of p35, p25, is capable of binding and activating Cdk5 as it contains the Cdk5 binding domain. Compared to Cdk5/p35, Cdk5/p25 is more stable and has a higher enzymatic activity both in vitro and in vivo [6, 7]. The first explanation is that p25 has a 3-fold longer half-life than p35 [8], which enables Cdk5 activity to last longer. Second, p25 has been shown to bind tighter to Cdk5 than p35 [9]. Third, Cdk5/p25 is more soluble and more freely accesses cytoplasmic and nuclear substrates, most of which are related to neurodegenerative diseases [7 , 11]. For example, the Cdk5/p25 complex causes hyperphosphorylation of tau and neurofilaments, leading to neurodegeneration and cell death [6, 12]. Elevated Cdk5/p25 has been found in the postmortem brains of Alzheimer’s disease (AD) [7], Parkinson’s disease (PD) [13 –15], and Huntington disease [16]. Aberrant Cdk5/p25 is correlated with neuronal death in animal models, including an amyotrophic lateral sclerosis mouse model [17, 18], a global cerebral ischemia mouse model [19], an Aβ-injected mouse model [20], and the 5x FAD mouse model [21]. Additionally, induced expression of p25 in the forebrain of transgenic (Tg) mice leads to hyperphosphorylated tau, neurofibrillary tangles, deposit of amyloid-β (Aβ) and neuroinflammation [22, 23], providing direct evidence that overexpression of p25 in the brain is neurotoxic. Therefore, p25 causes aberrant activation of Cdk5, which is closely connected to neurodegeneration.

Consequently, to inhibit Cdk5/p25 activity has been a treatment strategy for neurodegenerative diseases in preclinic studies. Roscovitine, a competitive inhibitor of the ATP-binding site of Cdks, provides inhibition of aberrant Cdk5. However, it may also have side effects on other Cdks [24]. Cdk5 inhibitory peptide (CIP), a fragment of the c-terminus of p35 (125 amino acid), is the first identified inhibitory peptide selectively inhibiting the activity of Cdk5/p25 [25, 26]. It has been reported that CIP can specifically inhibit the activity of Cdk5/p25 without affecting the activity of Cdk5/p35 nor Cdc2 both in transfected HEK cells and primary cortical neurons [27], leading to the reduction of the tau hyperphosphorylation caused by the Cdk5/p25 complex [26] and the Aβ_1 - 42-induced apoptosis of cortical neurons [25]. Furthermore, CIP was able to rescue against hyperphosphorylated tau and amyloid pathologies and cognitive decline caused by hyperactivity of Cdk5/p25 without affecting normal neurodevelopment in CIP-p25 tetra Tg mice, where CIP acted as a pre-treatment since p25 was induced at adult age [23]. In our previous report, administration of adeno-associated virus 9 (AAV9) CIP reduces the pathological changes correlated with the reversal of memory loss and anxiety-like behavior in amyloid precursor protein/presenilin 1 (APP/PS1) mice after the onset of pathological changes [28]. In addition, pretreatment with AAV9-CIP reduces dopaminergic (DA) neuronal loss and alleviates motor and anxiety-like symptoms in a 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine-probenecid (MPTP/p) induced PD mouse model [29]. Thus, both CIP treatment in APP/PS1 and PD mouse models indirectly suggests the role of p25 in the progression of neurodegenerative diseases.

In the present study, we first established a p25 overexpression mouse model by intracerebroventricular (i.c.v.) injection of AAV8-GFP-p25 virus, and then evaluated whether administration of AAV9-CIP one month later would provide neuroprotection against toxicity of p25 overexpression. We found that the overexpression of p25 in the mouse brain led to neuroinflammation and hyperphosphorylated tau, accompanied by cognitive deficits and increased anxiety-like behavior. Administration of CIP alleviated the pathological and behavioral changes induced by p25 toxicity in mice.

MATERIALS AND METHODS

Preparation of AAV8-GFP-p25 and AAV9-CIP-T2A-mCherry virus

Expression of pAAV-hSYN-GFP-p25 plasmid was described in a previous study [30]. CIP fragment was generated by PCR using plasmid pcDNA3.1-C-p35 as a template and inserted into plasmid pHBAAV-hSYN-T2A-mCherry. The resulting plasmid pHBAAV-hSYN-CIP-T2A-mCherry expresses mCherry and CIP separately driven by the human SYN-1 promoter. Sequencing of CIP in the plasmid was confirmed by Sanger sequencing. AAV8-GFP-p25 and AAV9-CIP-T2A-mCherry were prepared by Hanbio Biotechnology Company (Shanghai, China). Briefly, to generate the AAV8-GFP-p25 virus: pAAV-hSYN-GFP-p25, AAV helper plasmid, and AAV2/8 containing the AAV2 rep and AAV8 cap genes were transfected into HEK293T cells. For the AAV9-CIP-T2A-mCherry virus: pHBAAV-hSYN-CIP-T2A-mCherry, AAV helper plasmid, AAV2/9 containing AAV2 rep and AAV9 cap genes were transfected into HEK293T cells. After packing and purification, the titers of AAV virus were measured by quantitative PCR (qPCR). Concentration of AAV8-GFP-p25 (AAV8-p25 hereafter) or AAV9-CIP-T2A-mCherry (AAV9-CIP hereafter) was 2.5×10¹² or 3.20×10¹² vg/ml (vg: virus copy numbers of genome). The working concentration of virus was 1×10¹² vg/ml.

Administration of AAV virus into mouse brain

All animal experiments were approved by the Southern Medical University Committee on Animal Care. Male C57BL/6 mice (8 weeks old) were purchased from animal facility of the Southern Medical University (Guangzhou, China). Mice were housed under a 12-h light/dark cycle with standard mouse chow and water ad libitum. Temperature (22±1°C) and humidity (55–70%) were controlled. Eight-week-old mice were randomly divided into three groups as illustrated in Fig. 2A: control group, mice injected with AAV vector; p25/vector group, mice injected with AAV8-p25, then injected with AAV9 vector one month later; and p25/CIP group, mice injected with AAV8-p25, then injected with AAV9-CIP one month later. For each virus, a 2 μl volume was used by i.c.v. injection into the mouse brain with a stereotaxic apparatus (Model 68010, RWD company, China) as described previously [28].

Fig.1

Delivery of p25 or CIP in mouse brain by i.c.v. injection of AAV virus. A) Microscope image indicated that AAV expressing plasmids pAAV- hSYN -GFP or pAAV- hSYN -GFP -p25 expressed green fluorescent GFP or GFP fused protein in the transfected HEK293T cells. Western blot analysis confirmed expression of GFP (27 kDa) or GFP-p25 fusion protein (51 kDa). B) Microscope image indicated that AAV expression plasmids pHBAAV-hSYN-T2A-mCherry or pHBAAV-hSYN-CIP-T2A-mCherry expressed red fluorescent mCherry in the transfected HEK293T cells. C) AAV8-GFP-p25 was injected via i.c.v. into the right lateral ventricle at 8 weeks old. One month later, AAV9-CIP-T2A-mCherry was injected via i.c.v. into the same lateral ventricle. One week later, slices from the hippocampus of mouse brains were assessed for the fluorescent signal under the microscope. Green or red fluorescence indicated expression of fusion protein GFP-p25 or mCherry.

Behavioral assays

Active avoidance test and open field test were used to assess learning ability and anxiety-like behavior. All animal experiments were performed by two independent observers. Behavioral tests were performed when mice were 6 months old. Before conducting all experiments, the mice were habituated for 2 h in the testing room. All experiments were carried out between 8 am and 5 pm. All methods were previously described as in references [28, 31].

Active avoidance test

Active avoidance test was performed in a shuttle box which consists of two identical compartments separated by a wall with a hole on the bottom. Both compartments are equipped with overhead stimulus lights (conditioned stimulus) and a grid floor which delivered a foot shock (unconditioned stimulus). The test session continued for 4 consecutive days with 60 trials per day. Mice were to allowed 3 min to adjust to the environment in the box before the test. During testing, the mice were placed into one compartment, and if the mice escaped to another compartment when the light was on and no foot shock was given, an active avoidance response would be recorded. The total number of active avoidance responses was used to assess the learning and memory ability of mice.

Open field test

The open field was performed in an empty white chamber (40 cm×40 cm×40 cm). Mice were placed into the chamber facing a random corner and the overhead camera recorded the trace of the mouse in the chamber for 30 min. Locomotor activity was measured by the total distance traveled, and anxiety behavior was measured by the ratio of distance exploring in the center area per total distance in the chamber.

Western blot

Western blot analysis was performed routinely. The protein samples from brain extractions were prepared and separated on 10–12% SDS-polyacrylamide gel electrophoresis (PAGE) gels and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA). The membranes were incubated with the following primary antibodies (Abs) with a 1:1000 dilution: S199 (phospho-tau pSer199 rabbit monoclonal Ab, Thermo, Waltham, MA), AT180 (phospho-tau Thr231 mouse monoclonal Ab, Thermo), TAU-5 (tau mouse monoclonal Ab, Thermo), AT8 (Phospho-PHF (paired helical filament)-tau pSer202 + Thr205 mouse monoclonal Ab, Thermo), mouse anti-glial fibrillary acidic protein (GFAP, Millipore, Billerica, MA), and anti-β-actin (Proteintech, Chicago, IL). Membranes were then incubated with the following secondary Abs at a 1:5000 dilution: horseradish peroxidase-conjugated goat anti-mouse Ab or anti-rabbit IgG Ab (Santa Cruz Biotechnology Inc, Santa Cruz, CA).

Immunohistochemistry

Mouse brains were fixed in 4% paraformaldehyde and embedded in paraffin and then cut into 4 μm slices by Leica169 RM2016 vibratome (Leica, Heidelberg, Germany). The slices were incubated with following primary Abs: mouse anti-GFAP (1:1000 dilution, Millipore, Billerica, MA), goat anti-Iba1 (1:500 dilution, Thermo scientific), and mouse anti-Aβ_1 - 42 (β Amyloid 1-42 Ab, 1:1000 dilution, Thermo scientific) overnight at 4°C. Goat anti-mouse or donkey anti-goat secondary Abs (ZSGB-BIO, Beijing, China) were used. Images of immunohistochemistry were analyzed by independent investigators.

qPCR

The qPCR experiment was performed as described [31]. Total RNA were prepared from mouse brains using the Trizol reagent (Taraka, Tokyo, Japan). RNA was reverse transcribed into first-strand cDNA using a reverse transcriptase kit (Prime script RT Master Mix, Taraka). mRNA expression levels were analyzed by light cycles 480 detection system (Roche Applied Bioscience, Indianapolis) using SYBR Premix Ex TaqTM II (Takara). The following primers were used: IL-1β,5’-GGATGAG -GACATGAGCACCT-3’, 5’-AGCTCATATGGGTCCG- ACAG-3’; TNF-α, 5’-GCTGAGCTCAAACCCTGGTA-3’, 5’-CGGACTCCGCA- AAGTCTAAG-3’. Quantification of target genes was normalized with β-actin. 2^–ΔΔCt method was used to quantify the relative amounts of mRNA level expression.

Statistical analysis

The results of the active avoidance test were analyzed by repeated-measures ANOVA. All other data were analyzed using one-way ANOVA. Statistical analysis was performed by SPSS 20 (IBM, Armonk, NY) and GraphPad Prism 6.05 (GraphPad, La Jolla, CA). All values were presented as the mean of at least three determinations±SEM, and a p-value of <0.05 was considered significant.

RESULTS

Delivery of p25 and CIP sequentially in the brain of mice by AAV vectors

To overexpress p25 in the adult mouse brain, we first constructed an AAV-p25 expressing plasmid pAAV-hSYN-GFP-p25, in which p25 is in-frame with GFP at its c-terminus and is driven by the human synapsin-1 promoter to enable the expression of the target gene in neurons in vivo [32 –34]. As indicated in Fig. 1A, transfected pAAV-hSYN-GFP or pAAV-hSYN-GFP-p25 in HEK 293 cells expressed a green fluorescent signal. Western blot analysis confirmed the predicted protein size of GFP (27 kDa) or GFP-p25 (51 kDa). After viral packing and purification, AAV8-GFP or AAV8-GFP-p25 virus was i.c.v injected. Expression of GFP or GFP-p25 fusion protein (Fig. 1C) was verified.

Next, the CIP expressing plasmid pHBAAV-hSYN-CIP-T2A-mCherry was constructed, in which the CIP and mCherry fluorescent proteins were expressed separately but both driven by the hSYN promoter to ensure expression in neurons. Plasmid pHBAAV-hSYN-T2A-mCherry or pHBAAV-hSYN-CIP-T2A-mCherry was transfected in HEK293 cells and expression of target protein was confirmed by mCherry fluorescence under microscopy, as indicated in Fig. 1B. Due to the small size of CIP, expression of CIP was only confirmed by qPCR instead of western blot analysis (data not shown). To generate the AAV-CIP virus, the AAV9 vector was used in order to avoid immunological reaction, which is dependent on the capsid protein of AAV. One month after the AAV8-p25 virus was injected, AAV9-CIP was i.c.v administrated. AAV9 vector alone was injected as control. One week later, slices from the hippocampus and cortex of mice were assessed for fluorescent signal under a microscope. As shown in Fig. 1C, strong signal of green or red fluorescence was found in the hippocampus, indicating expression of GFP-p25 fusion protein or CIP (indirectly) respectively. Weak expression of GFP-p25 or CIP was found in the cortex close to the hippocampus (data not shown). These data suggest that AAV virus successfully delivered p25 or CIP into the mouse brain.

CIP reduces the level of hyperphosphorylated tau induced by overexpression of p25 in mice

Here we established a p25 overexpressing mouse model followed by CIP treatment. As indicated in Fig. 2A, all 8-wk old mice were randomly divided into the following groups: a control group, a p25/vector group, and a p25/CIP group. AAV8-p25 was i.c.v injected into the mice to establish a p25 overexpression mouse model, using the AAV8 vector was a control. One month later, the AAV9-CIP or AAV9-vector virus were i.c.v injected. At the end point of 6 months, brains tissues and histological slices were obtained after animal behavioral studies. Atrophy of brain was not observed.

Fig.2

CIP reduced p25-mediated tau phosphorylation. All the brain lysates and histological slices were prepared when mice were 6 months old. A) Illustration of animal experiment schedule. Control group: mice were injected with AAV8-GFP at 8 weeks old and AAV9-vector at 12 weeks old; p25/vector group: the mice were injected with AAV8-GFP-p25 at 8 weeks old and AAV9-vector at 12 weeks old; p25/CIP group: the mice were injected with AAV8-GFP-p25 at 8 weeks old and AAV9-CIP at 12 weeks old. All behavioral changes were assessed when mice were 6 months old. 2 μl of 10¹²vg/ml AAV virus was injected into the right lateral cerebral ventricle per mouse per time. B) Western blot analyses were performed with the Abs AT-180, AT-8 and Ser199 which indicates tau phosphorylation at sites Thr231, Ser202/Thr205, and Ser199, respectively. Levels of tau phosphorylation of Thr231, Ser202/Thr205and Ser199 increased in p25/vector group compared to the control group (AT-180 : 1.832±0.256, p = 0.003; AT-8:1.740±0.210, p = 0.02; Ser199:1.660±0.259, p = 0.012), TAU-5 was used as a loading control. CIP treatment decreased tau phosphorylation at sites Thr231 and Ser199 compared to p25/vector mice (AT-8:1.294±0.075 versus 1.740±0.210, p = 0.031; Ser199:1.160±0.086 versus 1.660±0.259, p = 0.044). Data are presented as mean±SEM, n = 5 per group, *p < 0.05, **p < 0.01, versus p25/vector group. C) Representative immunohistochemistry images of Aβ_1 - 42 staining from hippocampus of the three groups. No remarkable Aβ deposition was observed in any mouse.

Hyperphosphorylation of tau and amyloid accumulation are common pathological changes in AD patients. Tau hyperphosphorylation, microtubule polymerization, and Aβ deposition in the brain of p25Tg mice were reminiscent of AD [35], which can be ameliorated by transgenic expression of CIP [23]. Previously, systemic administration of AAV9-p25 by tail-vain injection induces tau phosphorylation on sites Ser199, Ser202/Thr205, and Thr231 without Aβ deposition in the cortex and hippocampus [30]. Therefore, tau phosphorylation status and Aβ deposition were analyzed in the present study. As showed in Fig. 2B, significantly increased levels of tau phosphorylation on sites Ser199, Thr231(by AT 180) and Ser202/Thr205 (by AT-8 Ab) were observed in the brain samples of AAV8-GFP-p25 mice, indicating tau hyperphosphorylation. Except for Thr231, phospho-tau Ser199 and Ser202/Thr205 were attenuated by AAV9-CIP treatment. Unlike p25Tg mice and in agreement with AAV9-GFP-p25 overexpression mice by systemic administration, no remarkable Aβ deposition appeared in AAV-GFP-p25 mice, as shown in Fig. 2C.

Together, these data suggest that administration of AAV9-CIP reduced the hyperphosphorylation of tau induced by overexpression of p25.

CIP inhibits neuroinflammation triggered by overexpression of p25 in mice

Neuroinflammation emerges as a key player in neurodegenerative diseases [36] and persistent inflammation leads to severe brain atrophy typical in the late phase of AD [37]. Indeed, early activation of astrocytes and then microglia have been clearly shown in p25Tg mice [35]. Similarly, increased staining of GFAP, a marker of astrocytes, was found in the hippocampus of AAV8-GFP-p25 mice by immunohistochemistry, which was reduced by CIP treatment, as shown in Fig. 3A and quantified as in Fig. 3B. Western blot analysis of brain samples confirmed the increased level of GFAP in p25 overexpression mice, which was significantly reduced in AAV8-GFP-p25/AAV9-CIP mice, as shown in Fig. 3C. However, activation of microglia in the hippocampus by Iba1 staining was not found (data no shown).

Fig.3

CIP treatment inhibited p25-induced inflammation in mice. All brain lysates and histological slices were prepared when mice in 6-month-old mice. A, B) Immunostaining of GFAP, a marker of activation of astrocytes, in the hippocampus. p25/vector mice showed increased GFAP staining compared to control mice (40.82±1.499 versus 26.44±2.337, p = 0.002). GFAP staining was inhibited by CIP treatment (32.26±2.788 versus 40.82±1.499, p = 0.026). n = 4 per group. C) Western blot analysis. The level of GFAP in total brain lysates significantly increased in p25/vector mice than that of control mice (1.976±0.327, p = 0.004), while CIP treatment decreased expression of GFAP, compared to p25/vector mice (1.334±0.081 versus 1.976±0.327, p = 0.038). β-actin was used as the loading control. n = 5 per group. D) Inflammatory cytokines IL-1 and TNF-α were detected by qPCR. IL-1 and TNF-α both showed higher expression levels in p25/vector mice, compared to control mice (IL-1:1.584±0.215, p = 0.037; TNF-α: 2.874±0.862, p = 0.021). CIP treatment reduced expression of IL-1 and TNF-α (IL-1:0.897±0.225 versus 1.584±0.215, p = 0.016; TNF-α: 0.821±0.229 versus 2.874±0.862, p = 0.013). n = 5 per group. All the data are presented as mean±SEM, *p < 0.05, **p < 0.01, versus p25/vector group.

In addition, expression of IL-1 and TNF-α, two common cytokines related to inflammation, were further analyzed with qPCR. Compared with control mice, the mRNA levels of IL-1 and TNF-α were significantly elevated in AAV8-GFP-p25 mice and were significantly reduced in AAV8-GFP-p25/AAV9-CIP mice, as showed in Fig. 3D (*p < 0.05).

Taken together, the overexpression of p25 in mice triggers inflammation in the brains of mice, which was attenuated by CIP treatment.

AAV9-CIP treatment ameliorates behavioral changes in AAV8-GFP-p25 overexpression mice

Increased levels of the p25 protein have been found in the brain of postmortem AD patients [7]. To study the toxicity of p25 in neurons, several p25 overexpression mouse models were established. Significant memory impairment was further confirmed in p25Tg mice [23] and p25 overexpression mice via AAV9-GFP-p25 injection through tail-vein [30]. Importantly, transgenic expression of CIP in p25-Tg mice reverses cognitive deficits, in which CIP expression is 6 weeks earlier than p25 induction [23].

In the present study, we would like to evaluate whether CIP treatment after the overexpression of p25 one month later would have protective effects. Behavioral changes were assessed at three month later after CIP treatment. The active avoidance response test was used to evaluate learning and memory ability. As shown in Fig. 4A, p25 overexpression mice had significantly less active avoidance responses since day 2 compared with AAV vector control (*p < 0.05), which was increased after CIP treatment since day 2 and was significantly different at day 4 (^#p < 0.05). In addition, the open field assay, which evaluates locomotor activity and anxiety-like behavior, was used. As in Fig. 4B, p25 overexpression mice had travelled less distance and preferred to spend more time along the sides of the chamber. Compared to vector control mice, p25 overexpression mice traveled more total distance (**p < 0.01) and less distance in the center (*p < 0.05). Although there was a tendency of CIP treatment to improve the anxiety-like behavior as shown by the traveling track, there was no difference between the AAV8-GFP-p25 group and AAV8-GFP-p25/AAV9-CIP group in total travel distance and the ratio of center per total distance.

Fig.4

Behavioral alteration in AAV8-p25 injected mice was partially restored by AAV9-CIP treatment. A) Active avoidance tests were performed between the three groups. The p25/vector mice failed to show active avoidance learning in contrast to the control mice starting at day 2 (20.500±3.964, n = 6 versus 33.450±3.599, n = 11; p = 0.031). p25/CIP mice significantly reversed the active avoidance response on day 4 compared to p25/vector mice (36.500±2.29, n = 10 versus 24.160±3.833, n = 6; p = 0.011). B) An illustrative example of one mouse’s travel pathway per group in the Open Field Test. The p25/vector mice showed significantly less activity than the control mice as measured both in the ratio of center/total distance (0.129±0.014, n = 6 versus 0.218±0.156, n = 8; p = 0.04) and in the total distance traveled (7413±536.1 cm, n = 6 versus 10147±562.7 cm, n = 8; p = 0.001). Although there was a tendency for more activity in p25/CIP mice than in p25/vector mice, no significant difference was found between two groups. All the data are presented as mean±SEM, *p < 0.05, **p < 0.01, control group versus p25/vector group; ^#p < 0.05, p25/CIP group versus p25/vector group.

These data indicate that the overexpression of p25 in neurons impaired learning and memory ability and increased anxiety-like behavior in mice. Importantly, the administration of CIP delivered by AAV impedes the learning and memory ability deficit caused by p25.

DISCUSSION

Aberrant Cdk5 activity induced by p25 causes hyperphosphorylation of many downstream substrates related to the pathogenesis of degenerative diseases including AD. The first evidence is from the observation that increased activity of Cdk5/p25 is related to the pathological changes in the postmortem brains of AD [7]. More studies have demonstrated that hyperactivity of Cdk5/p25 is involved in neuronal death in degenerative disease models, such as amyotrophic lateral sclerosis [17, 18] and AD [20, 21]. Transgenic expression of p25 causes hyperphosphorylated tau, neurofibrillary tangles, Aβ deposition, and neuroinflammation [22, 23]. Therefore, inhibiting the activity of Cdk5/p25 is a potential treatment strategy. Pre-expression of CIP prevents the toxicity of induced p25 in CIP-p25 tetra-Tg mice [23]. In the present study, we established an alternative p25 overexpression mouse model by i.c.v injection of the AAV-GFP-p25 virus, which characterized by hyperphosphorylation of tau and inflammation accompanied by impaired learning ability. After administration of AAV-CIP one-month post p25 insult in mice, the pathological changes as well as behavioral abnormality were reversed.

To investigate the mechanism of p25 toxicity in vivo, several p25 overexpression mouse models have been established. Transgenic mice with different promoters to drive expression of p25 exhibit various severity of pathological and behavioral changes. The first p25 transgenic mouse overexpressing p25 driven by the rat neuron-specific enolase (NSE) promoter, presented with whole-body exertion tremors at 4–9 weeks of age, accompanying hyperphosphorylation of tau, and neurofilaments in the amygdala, hypothalamus, and cortex [38]. The second p25Tg mouse model expressing p25 in an inducible manner in the excitatory neurons of the forebrain by the application of the tetracycline-controlled transactivator (tTA) and CaMKII-promoter. This mouse model recapitulates many features of AD-like pathological changes including neuronal loss, astrocytosis, NFTs, and Aβ deposition [22, 23]. A similar constructive strategy was used to establish another p25Tg mouse line by an independent laboratory, which only showing hippocampal sclerosis, neocortical degeneration, and activated microglia but no tau phosphorylation and apoptosis [37]. Although different p25Tg mice are good tools for studying the role of p25, p25Tg mice are hard to breed and require tedious procedures and time. To solve this problem, we applied the AAV virus system, which is currently widely used as a gene delivery system due to its non-pathogenicity and high efficiency in delivering the target gene into different cell types [39]. Among all serotypes [40], AAV8, AAV9, AAV10, and AAV11 are able to cross the blood-brain barrier [41, 42]. We have previously set up an AAV9-mediated p25-overexpression mouse model by tail-vein injection, which presented as hyperphosphorylated tau, NFT formation, astrocytosis, microgliosis, and impaired learning ability [30]. To avoid systemic effects, we i.c.v. injected the AAV8-GFP-p25 virus instead of injection by tail-vein. These p25 overexpression mice showed hyperphosphorylation of tau and inflammation in the hippocampus accompanied by impairment of learning and memory. Our results indicated that AAV-mediated p25 overexpression mice is an alternative model to study the toxicity of p25 even though the model has a milder phenotype than p25Tg mice.

NFTs due to hyperphosphorylated tau is one of the pathological hallmarks of AD [43]. Many phosphorylation sites on tau by different kinases have been found in AD. It has been shown that Cdk5 phosphorylates tau at sites Thr181, Ser199, Ser202, Thr205, Thr212, Ser214, Thr217, Thr231, Ser235, Ser396, and Ser404, but not at Ser262, Ser400, Thr403, Ser409, Ser413, or Ser422 [44]. Above all of them, Ser199, Ser202/Thr205, and Thr231 are probably the most active sites related to Cdk5 activity, which may be modulated by other kinases, such as cAMP-dependent protein kinase [45], glycosidase [44], and quercetin [46]. It is noted that the Cdk5/p25 complex hyperphosphorylates many epitopes of tau such as Ser202, Thr205, Ser235, and Ser404 in AD [47], resulting in hyperphosphorylation of neurofilaments and tangle formation [48]. Here, we observed significantly increased levels of tau phosphorylation on Ser199, Ser202/Thr205 (by AT-8 Ab), and Thr231(by AT 180) in the brains of AAV8-GFP-p25 mice. After being treated with AAV9-CIP, levels of phospho-tau Ser199 and Ser202/Thr205 were attenuated but no change was observed in the level of phospho-tau Thr231, which could be due to the expression level of CIP at the checkpoint time. In transgenic CIP in the tetra Tg mice [23], consecutively expression of CIP inhibits phosphorylation on Ser202/Thr205 and Thr231. In our previous study in the APP/PS1 model [28], after i.c.v. injection of AAV9-GFP-CIP virus, although different from the vector used in the present study, inhibition on all three sites was observed after 12 weeks of treatment. Therefore, we assumed that the inhibition of these three phosphorylation sites on tau is dependent on the injection time of virus, the lower expression level, the longer injection duration. There are no arguments that tau phosphorylation and the consequent oligomerization play an important role in AD pathogenesis. Tau phosphorylation has been suggested as a primary trigger for tau aggregation, which ranges from soluble tau oligomers to insoluble tau deposits found in dystrophic neurites, neuropil threads, and NFTs in cell bodies. It has been shown that tau hyperphosphorylation promotes tau to behave abnormally, resulting in microtubule assembly disruption as well as a breakdown in tau self-assembly, and allowing tau to bind ATP to induce tau self-assembly into filaments [49]. However, microtubule affinity regulating kinase (MARK) and protein kinase A (PKA) phosphorylates tau at certain sites, including Ser262, Ser324, Ser356, and Ser320, rendering tau refractory to aggregation. Thus, phosphorylation itself is not sufficient to drive aggregation, but may help enhance ongoing multimerization events [50]. In this study, although tau phosphorylation was observed, obvious NFTs formation was not found (data not shown).

The Cdk5/p25 complex plays an important role in deposition of senile plaques composing of Aβ peptides [43]. Cdk5 increases Aβ production by carrying out the phosphorylation of amyloid protein precursor (APP) at Thr668 which leads to the β-secretase cleavage of APP [51, 52]. Aβ deposition has been found in p25Tg mice [22, 23] but is absent in mice with systemic administration of AAV9-p25 [30] or in mice with local injection of AAV8-p25 in the brains of mice in the present study. These data suggest that p25Tg mice had more severe pathological changes than AAV-mediated p25 overexpression mice.

Neuroinflammation is an additional causative factor alongside the neurodegenerative process. Astrocytes were confirmed to be activated in AD patients [53], and the activation of astrocytes induced the production of pro-inflammatory cytokines [54]. In the present study, we observed increased GFAP and proinflammatory cytokine levels in p25 overexpression mice. According to the data from p25Tg mice, astrogliosis is an early event, which is triggered by the upregulation of cytosolic phospholipase 2 and occurs much earlier (one week) before hyperphosphorylation of tau (4 weeks), microgliosis (4 weeks), and amyloid pathology (8 weeks) [35]. In the present study, we did not observe microgliosis and amyloid pathology, which again indicates that the pathological changes in our p25 overexpression mice is milder than that of p25Tg mice. Several studies indicate that neuroinflammation can directly modulate tau phosphorylation. Activation of microglia with lipopolysaccharides or amyloid precursor protein fragments induced neuronal tau phosphorylation in co-cultures with rat primary neocortical neurons [55]. Tumor necrosis factor-α (TNFα), another pro-inflammatory cytokine, also caused an increase in tau hyperphosphorylation [56]. An intra-hippocampal injection of 10 μg of LPS resulted in an increase in tau phosphorylation for several days [57]. Inflammatory stimuli, such as Aβ, stimulates microglial production of pro-inflammatory mediators such as IL-1β leading to the up-regulation of kinases involved in tau phosphorylation and exacerbation of the pathology. However, inflammation may benefit tau pathology by inducing microglial phagocytosis of extracellular tau species [58]. Altogether, the outcome of inflammation on tau pathology is, at least in part, dependent on the nature of immunological stressor and the pro-inflammatory cytokine involved. In the present study, we speculated phosphorylation of tau may be increased by inflammation and Cdk5/p25 hyperactivity. However, we did not check whether astrocytosis occurred earlier than hyperphosphorylation of tau in the present study.

Due to the pathological role of Cdk5/p25 in the process of neurodegenerative of AD, targeting Cdk5/p25 is a potential treatment strategy. Application of several p35 derivatives, CIP, P5, and TFP5 in disease models have been previously tested. CIP has been reported to have a much higher affinity for Cdk5 than p25 and is identified as a potential inhibitory peptide selective for Cdk5/p25 without affecting Cdk5/p35 activity [23 , 26]. Zheng et al have demonstrated that CIP markedly inhibits the activity of Cdk5/p25 in vitro and in transfected cells [26] Further, they found that CIP can specifically inhibit Cdk5/p25-induced tau hyperphosphorylation in neurons and Aβ_1 - 42-induced apoptosis. However, CIP did not inhibit endogenous or transfected p35/Cdk5 activity [25]. In the tetra Tg mice, Sundaram et al found that long-time transgenic expression of CIP had no effect on normal Cdk5/p35 or Cdk5/p39 activity [23]. Currently, the mechanism is still unclear. Truncated mutation study found that E240 in p35 is important in sustaining Cdk5 kinase activity [59]. Using crystal structure analysis and comparative dynamics stimulation, CIP, compared to p25, showed enhanced flexibility and large structural changes which repositions the residue E240 to be fully exposed to the solvent, leading to a higher binding affinity to Cdk5, which in turn inhibits the activity of Cdk5 kinase [60]. However, evidence is still lacking for a similar study comparing the binding affinity of CIP and p35 to Cdk5.

In p25 Tg mice, pre-expression of CIP counteracted the hyperphosphorylation of tau, neuroinflammation, and behavioral impairment induced by p25 [23]. We have previously utilized AAV9-CIP as pre-treatment in PD mice [29] or post-treatment after the onset of pathological changes in APP/PS1 mice with the ideal outcome [28]. P5 (Lys²⁴⁵–Ala²⁷⁷of p35) is another peptide derived from p35 cleavage that inhibited Cdk5/p25 activity [61]. The transmembrane peptide Tet modifying TP5 or TFP5 to enhance across the blood-brain-barrier demonstrates neuronal protection in AD or PD mouse models by repeated injections. However, AAV mediated CIP is more feasible for clinical applications than repeated injection of TP5. In conclusion, administration of AAV8-GFP-p25 in mice by i.c.v. injection generated a neurodegenerative-like mouse model which is useful for studying the functions of p25 in vivo. In addition, our data provides evidence that CIP may become a potential treatment strategy for neurodegenerative diseases.

Footnotes

ACKNOWLEDGMENTS

The work was supported by the National Nature, Science Fund of China (#81271430), Guangdong provincial scientific & technologic progression funds (# 2015B050501006, 2016A020215182), Guangdong Shunde regional scientific & technologic progression fund (2015CXTD06). The language of the manuscript is edited by Melanie Zhang, Department of Neurobiology Northwestern University, Feinberg School of Medicine, Evanston, IL 60208.

Authors’ disclosures available online ().

References

Lew

, Beaudette

, Litwin

, Wang

(1992) Purification and characterization of a novel proline-directed protein kinase from bovine brain. J Biol Chem 267, 13383–13390.

Tang

, Yeung

, Lee

, Matsushita

, Matsui

, Tomizawa

, Hatase

, Wang

(1995) An isoform of the neuronal cyclin-dependent kinase 5 (Cdk5) activator. J Biol Chem 270, 26897–26903.

, Tsai

(2011) Cyclin-dependent kinases in brain development and disease. Annu Rev Cell Dev Biol 27, 465–491.

Dhavan

, Tsai

(2001) A decade of CDK5. Nat Rev Mol Cell Biol 2, 749–759.

Nikolic

, Dudek

, Kwon

, Ramos

, Tsai

(1996) The cdk5/p35 kinase is essential for neurite outgrowth during neuronal differentiation. Genes Dev 10, 816–825.

Lee

, Kwon

, Li

, Peng

, Friedlander

, Tsai

(2000) Neurotoxicity induces cleavage of p35 to p25 by calpain. Nature 405, 360–364.

Patrick

, Zukerberg

, Nikolic

, de la Monte

, Dikkes

, Tsai

(1999) Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Nature 402, 615–622.

Patrick

, Zhou

, Kwon

, Howley

, Tsai

L-H

(1998) p35, the neuronal-specific activator of cyclin-dependent kinase 5 (Cdk5) is degraded by the ubiquitin-proteasome pathway. J Biol Chem 273, 24057–24064.

Amin

, Albers

, Pant

(2002) Cyclin-dependent kinase 5 (cdk5) activation requires interaction with three domains of p35. J Neurosci Res 67, 354–362.

10.

Asada

, Yamamoto

, Gohda

, Saito

, Hayashi

, Hisanaga

(2008) Myristoylation of p39 and p35 is a determinant of cytoplasmic or nuclear localization of active cyclin-dependent kinase 5 complexes. J Neurochem 106, 1325–1336.

11.

Kusakawa

, Saito

, Onuki

, Ishiguro

, Kishimoto

, Hisanaga

(2000) Calpain-dependent proteolytic cleavage of the p35 cyclin-dependent kinase 5 activator to p25. J Biol Chem 275, 17166–17172.

12.

Noble

WOV

, Takata

, Casey

, Mary

, Meyerson

, Gaynor

, LaFrancois

, Wang

, Kondo

, Davies

, Burns

, Veeranna , Nixon

, Dickson

, Matsuoka

, Ahlijanian

, Lau

, Duff

(2003) Cdk5 is a key factor in tau aggregation and tangle formation in vivo. Neuron 38, 555–565.

13.

Brion

, Couck

(1995) Cortical and brainstem-type Lewy bodies are immunoreactive for the cyclin-dependent kinase 5. Am J Pathol 147, 1465–1476.

14.

Takahashi

, Iseki

, Kosaka

(2000) Cyclin-dependent kinase 5 (Cdk5) associated with Lewy bodies in diffuse Lewy body disease. Brain Res 862, 253–256.

15.

Nakamura

, Kawamoto

, Nakano

, Akiguchi

, Kimura

(1997) p35nck5a and cyclin-dependent kinase 5 colocalize in Lewy bodies of brains with Parkinson’s disease. Acta Neuropathol 94, 153–157.

16.

Paoletti

, Vila

, Rife

, Lizcano

, Alberch

, Gines

(2008) Dopaminergic and glutamatergic signaling crosstalk in Huntington’s disease neurodegeneration: The role of p25/cyclin-dependent kinase 5. J Neurosci 28, 10090–10101.

17.

Nguyen

, Julien

(2003) Cyclin-dependent kinase 5 in amyotrophic lateral sclerosis. Neurosignals 12, 215–220.

18.

Nguyen

, Lariviere

, Julien

(2001) Deregulation of Cdk5 in a mouse model of ALS: Toxicity alleviated by perikaryal neurofilament inclusions. Neuron 30, 135–147.

19.

Castro-Alvarez

, Gutierrez-Vargas

, Darnaudery

, Cardona-Gomez

(2011) ROCK inhibition prevents tau hyperphosphorylation and p25/CDK5 increase after global cerebral ischemia. Behav Neurosci 125, 465–472.

20.

Lopes

, Oliveira

, Agostinho

(2010) Neurodegeneration in an Abeta-induced model of Alzheimer’s disease: The role of Cdk5. Aging Cell 9, 64–77.

21.

Shukla

, Zheng

, Mishra

, Amin

, Steiner

, Grant

, Kesavapany

, Pant

(2013) A truncated peptide from p35, a Cdk5 activator, prevents Alzheimer’s disease phenotypes in model mice. FASEB J 27, 174–186.

22.

Cruz

, Tseng

, Goldman

, Shih

, Tsai

(2003) Aberrant Cdk5 activation by p25 triggers pathological events leading to neurodegeneration and neurofibrillary tangles. Neuron 40, 471–483.

23.

Sundaram

, Poore

, Sulaimee

, Pareek

, Asad

, Rajkumar

, Cheong

, Wenk

, Dawe

, Chuang

, Pant

, Kesavapany

(2013) Specific inhibition of p25/Cdk5 activity by the Cdk5 inhibitory peptide reduces neurodegeneration in vivo. J Neurosci 33, 334–343.

24.

Meijer

, Borgne

, Mulner

, Chong

, Blow

, Inagaki

, Delcros

, Moulinoux

(1997) Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur J Biochem 243, 527–536.

25.

Zheng

, Kesavapany

, Gravell

, Hamilton

, Schubert

, Amin

, Albers

, Grant

, Pant

(2005) A Cdk5 inhibitory peptide reduces tau hyperphosphorylation and apoptosis in neurons. EMBO J 24, 209–220.

26.

Zheng

, Li

, Amin

, Albers

, Pant

(2002) A peptide derived from cyclin-dependent kinase activator (p35) specifically inhibits Cdk5 activity and phosphorylation of tau protein in transfected cells. Eur J Biochem 269, 4427–4434.

27.

Kesavapany

, Zheng

, Amin

, Pant

(2007) Peptides derived from Cdk5 activator p35, specifically inhibit deregulated activity of Cdk5. Biotechnol J 2, 978–987.

28.

, Pan

, Xu

, He

, Huang

, Song

, Huang

, Zhang

, Hu

(2017) Adeno-associated viral 9-mediated Cdk5 inhibitory peptide reverses pathologic changes and behavioral deficits in the Alzheimer’s disease mouse model. FASEB J 31, 3383–3392.

29.

, Huang

, Xu

, Song

, Huang

, Xie

, Hu

(2018) Cdk5 inhibitory peptide prevents loss of dopaminergic neurons and alleviates behavioral changes in an MPTP induced Parkinson’s disease mouse model. Front Aging Neurosci 10, 162.

30.

Zhou

, Huang

, Pan

, Xu

, He

, Ji

, Hu

(2016) Neurodegeneration-like pathological and behavioral changes in an AAV9-mediated p25 overexpression mouse model. J Alzheimers Dis 53, 843–855.

31.

, Song

, Huang

, He

, Zhou

, Gu

, Pan

, Hu

(2018) Disruption of AT-hook 1 domain in MeCP2 protein caused behavioral abnormality in mice. Biochim Biophys Acta 1864, 347–358.

32.

Marro

, Pang

, Yang

, Tsai

, Qu

, Chang

, Sudhof

, Wernig

(2011) Direct lineage conversion of terminally differentiated hepatocytes to functional neurons. Cell Stem Cell 9, 374–382.

33.

Son

, Ichida

, Wainger

, Toma

, Rafuse

, Woolf

, Eggan

(2011) Conversion of mouse and human fibroblasts into functional spinal motor neurons. Cell Stem Cell 9, 205–218.

34.

Pang

, Yang

, Vierbuchen

, Ostermeier

, Fuentes

, Yang

, Citri

, Sebastiano

, Marro

, Sudhof

, Wernig

(2011) Induction of human neuronal cells by defined transcription factors. Nature 476, 220–223.

35.

Sundaram

, Chan

, Poore

, Pareek

, Cheong

, Shui

, Tang

, Low

, Wenk

, Kesavapany

(2012) Cdk5/p25-induced cytosolic PLA2-mediated lysophosphatidylcholine production regulates neuroinflammation and triggers neurodegeneration. J Neurosci 32, 1020–1034.

36.

Dorothee

(2018) Neuroinflammation in neurodegeneration: Role in pathophysiology, therapeutic opportunities and clinical perspectives. J Neural Transm (Vienna) 125, 749–750.

37.

Muyllaert

, Terwel

, Kremer

, Sennvik

, Borghgraef

, Devijver

, Dewachter

, Van Leuven

(2008) Neurodegeneration and neuroinflammation in cdk5/p25-inducible mice: A model for hippocampal sclerosis and neocortical degeneration. Am J Pathol 172, 470–485.

38.

Ahlijanian

, Barrezueta

, Williams

, Jakowski

, Kowsz

, McCarthy

, Coskran

, Carlo

, Seymour

, Burkhardt

, Nelson

, McNeish

(2000) Hyperphosphorylated tau and neurofilament and cytoskeletal disruptions in mice overexpressing human p25, an activator of cdk5. Proc Natl Acad Sci U S A 97, 2910–2915.

39.

Grieger

, Samulski

(2005) Adeno-associated virus as a gene therapy vector: Vector development, production and clinical applications. Adv Biochem Eng Biotechnol 99, 119–145.

40.

Day

, Byrne

, Schaffer

, Flannery

(2014) Advances in AAV vector development for gene therapy in the retina. Adv Exp Med Biol 801, 687–693.

41.

Iida

, Takino

, Miyauchi

, Shimazaki

, Muramatsu

(2013) Systemic delivery of tyrosine-mutant AAV vectors results in robust transduction of neurons in adult mice. Biomed Res Int 2013, 974819.

42.

Gombash

, Cowley

, Fitzgerald

, Hall

, Mueller

, Christofi

, Foust

(2014) Intravenous AAV9 efficiently transduces myenteric neurons in neonate and juvenile mice. Front Mol Neurosci 7, 81.

43.

Bhounsule

, Bhatt

, Prabhavalkar

, Oza

(2017) Cyclin dependent kinase 5: A novel avenue for Alzheimer’s disease. Brain Res Bull 132, 28–38.

44.

Liu

FIK

, Grundke-Iqbal

, Gong

(2002) Involvement of aberrant glycosylation in phosphorylation of tau by cdk5 and GSK-3β. FEBS Lett 530, 209–214.

45.

Liu

, Liang

, Shi

, Yin

, El-Akkad

, Grundke-Iqbal

, Iqbal

, Gong

(2006) PKA modulates GSK-3beta- and cdk5-catalyzed phosphorylation of tau in site- and kinase-specific manners. FEBS Lett 580, 6269–6274.

46.

Shen

, Luo

, Li

, Ting

, He

, Xu

, Wang

(2018) Quercetin inhibits okadaic acid-induced tau protein hyperphosphorylation through the Ca2+calpainp25CDK5 pathway in HT22 cells. Int J Mol Med 41, 1138–1146.

47.

Kimura

, Ishiguro

, Hisanaga

(2014) Physiological and pathological phosphorylation of tau by Cdk5. Front Mol Neurosci 7, 65.

48.

Zhou

, Wang

, Feng

, Chen

(2010) Increased expression of cdk5/p25 in N2a cells leads to hyperphosphorylation and impaired axonal transport of neurofilament proteins. Life Sci 86, 532–537.

49.

Farid

, Corbo

, Alonso Adel

(2014) Tau binds ATP and induces its aggregation. Microsc Res Tech 77, 133–137.

50.

, Pina-Crespo

, Zhang

, Chen

, Xu

(2017) Tau-mediated neurodegeneration and potential implications in diagnosis and treatment of Alzheimer’s disease. Chin Med J (Engl) 130, 2978–2990.

51.

Liu

, Wang

, Jiang

, Tan

, Xing

, Yu

(2016) The role of Cdk5 in Alzheimer’s disease. Mol Neurobiol 53, 4328–4342.

52.

Song

, Son

, Lee

, Seo

, Kim

, Chung

(2015) Enhancement of BACE1 activity by p25/Cdk5-mediated phosphorylation in Alzheimer’s disease. PLoS One 10, e0136950.

53.

Rao

, Rapoport

, Kim

(2011) Altered neuroinflammatory, arachidonic acid cascade and synaptic markers in postmortem Alzheimer’s disease brain. Transl Psychiatry 1, e31.

54.

Steele

, Robinson

(2012) Reactive astrocytes give neurons less support: Implications for Alzheimer’s disease. Neurobiol Aging 33, 423e421–413.

55.

Li Y

, Barger

, Griffin

(2003) Interleukin-1 mediates pathological effects of microglia on tau phosphorylation and on synaptophysin synthesis in cortical neurons through a p38-MAPK pathway. J Neurosci 23, 1605–1611.

56.

Janelsins

, Mastrangelo

, Park

, Sudol

, Narrow

, Oddo

, LaFerla

, Callahan

, Federoff

, Bowers

(2008) Chronic neuron-specific tumor necrosis factor-alpha expression enhances the local inflammatory environment ultimately leading to neuronal death in 3xTg-AD mice. Am J Pathol 173, 1768–1782.

57.

Lee

, Rizer

, Selenica

, Reid

, Kraft

, Johnson

, Blair

, Gordon

, Dickey

, Morgan

(2010) LPS- induced inflammation exacerbates phospho-tau pathology in rTg4510 mice. J Neuroinflammation 7, 56.

58.

Barron

, Gartlon

, Dawson

, Atkinson

, Pardon

(2017) A state of delirium: Deciphering the effect of inflammation on tau pathology in Alzheimer’s disease. Exp Gerontol 94, 103–107.

59.

Tang

, Chun

, Zhang

, Wang

(1997) Cyclin-dependent kinase 5 (Cdk5) activation domain of neuronal Cdk5 activator. Evidence of the existence of cyclin fold in neuronal Cdk5a activator. J Biol Chem 272, 12318–12327.

60.

Cardone

, Hassan

, Albers

, Sriram

, Pant

(2010) Structural and dynamic determinants of ligand binding and regulation of cyclin-dependent kinase 5 by pathological activator p25 and inhibitory peptide CIP. J Mol Biol 401, 478–492.

61.

Zheng

, Amin

, Hu

, Rudrabhatla

, Shukla

, Kanungo

, Kesavapany

, Grant

, Albers

, Pant

(2010) A 24-residue peptide (p5), derived from p35, the Cdk5 neuronal activator, specifically inhibits Cdk5-p25 hyperactivity and tau hyperphosphorylation. J Biol Chem 285, 34202–34212.